Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Image segmentation method converting unsupervised cluster into self-supervised classification

A technology for supervised classification and image segmentation, which is applied in image analysis, image data processing, instrumentation, etc., and can solve problems such as unacceptable segmentation results.

Inactive Publication Date: 2015-01-21
SHANGHAI DIANJI UNIV
View PDF2 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Although some research results have been achieved in texture image segmentation based on pixel feature clustering methods, the segmentation results are sometimes unacceptable due to the diversity and irregularity of textures.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Image segmentation method converting unsupervised cluster into self-supervised classification
  • Image segmentation method converting unsupervised cluster into self-supervised classification
  • Image segmentation method converting unsupervised cluster into self-supervised classification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] In order to make the content of the present invention clearer and easier to understand, the content of the present invention will be described in detail below in conjunction with specific embodiments and accompanying drawings.

[0025] The existing method is to cluster the features without any training samples, that is, to use the so-called unsupervised learning method. Since there is no guidance from the training samples, the robustness of the segmentation results is significantly lower than that obtained by using a classifier trained on typical samples. Therefore, we can try to find some typical training samples to guide the segmentation, and change the original unsupervised learning method into a self-supervised learning method, thereby improving the robustness of the algorithm.

[0026] Moreover, in the segmentation process, it is usually assumed that all sample features belong to the same Gaussian distribution, and there is no division of subspaces. When the real ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides an image segmentation method converting an unsupervised cluster into a self-supervised classification. The method comprises the flowing sequentially executed steps that firstly, specific samples are extracted from an image and used as training samples through a fuzzy C-means method according to the characteristics of the samples, and an unsupervised clustering problem is converted into a self-supervised classification problem; secondly, subspace partition is carried out on the samples on the basis that the unsupervised clustering problem is converted into the self-supervised classification problem, and therefore an independent subspace is generated for each class of the self-supervised classification; thirdly, category classification is carried out on other samples except the specific samples in an iteration process with the help of KL conversion of multiple subspaces on the basis of the partitioned subspaces.

Description

technical field [0001] The invention relates to an image segmentation method for converting non-supervised clustering into self-supervised classification. Background technique [0002] There are two types of image segmentation: complete segmentation and partial segmentation. Complete segmentation has a great impact on the subsequent processing of images, so the requirements are relatively high. However, full segmentation is based on partial segmentation, which is a means of forming homomorphic regions for a certain feature, which do not directly correspond to objects in the image. In order to obtain a high-level complete segmentation from an image, partial segmentation can be performed first, and then based on it, high-level knowledge related to image content can be used for processing. Segmentation based on texture features belongs to the partial segmentation method in image segmentation. It uses texture features that can better describe the local pixel distribution struct...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06T7/00G06K9/66
Inventor 胡静
Owner SHANGHAI DIANJI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products